The course shows how a causal theory can be represented by a path diagram and translated into a structural equation model. We will deal with measurement models relating single or multiple indicators to latent variables. Different specifications of measurement models are tested via confirmatory factor analysis(CFA) as a submodel of SEM. Special emphasis is given to the use of such models to measure values and attitudes, the differentiation between formative and reflective indicators and the use of Multiple Group Confirmatory Factor Analysis(MGCFA) to study invariance of measurement over groups and nations. In the second part we will expand these models to take into account structural relations between latent variables and measurement models simultaneously.
Special topics include second and higher-order confirmatory factor analysis, Mediation and Moderation, MIMIC Models with formative and reflective indicators and the use of multiple groups SEM to study differences over time and nations and the topic of missing values. Special attention is given to the process of model modification and testing for invariance over groups and cultures using data from the European Social Survey (ESS) and the International Social Survey Program( ISSP) both in the presentations and the exercises. As Software we use the MPLUS Program. Theory and practical work with prepared data-sets will be conducted interchangeably. Participants are encouraged to bring additionally their own data. Time will also be dedicated for testing one`s own models. Participants are expected to attend computer labs daily and encouraged at the end to present their own analyses.
encouraging the training, research and cross-national co-operation of political scientists
